In this project we explore the pre-clinical application of the concept of Adaptive Therapy (AT), a term coined by the proposers of this grant. AT shifts the treatment paradigm of currently incurable cancers from the maximum tumor cell killing (all-out attack) approach to the use of evolutionarily enlightened drug combinations to stabilize the tumor burden. Our goal is to minimize the probability of emergence of multi drug resistant clones (MDR), which are the ultimate cause of current cancer treatment failure and patient death. In addition to the original AT idea of promoting competition of chemoresistant and chemosensitive tumor cells, the authors propose to explore evolutionary-based approaches that increase the cost of chemoresistance, making these cells less fit than their chemosensitive counterpart, and thus reducing the likeliness of emergence of a resistant tumor. Finally, the authors propose the use of clinically available non-invasive metabolic imaging techniques (FdG PET -Positron Emission Tomography) to assess tumor chemoresistance, and use this to inform the clinicians about the presence and number of resistant cells within the tumor. Our proof-of-concept tumor model will be in breast cancer. We will use in vitro and mouse models, non- invasive imaging of tumor burden and metabolism, and computational evolutionary models of tumor progression and drug response. These approaches will be used to detect and measure tumor resistance and suggest the optimal treatment strategy that maximally prolongs the duration of response to conventional breast cancer therapy. The authors propose to combine three chemotherapeutic strategies currently used in the clinic: (a) hormonal therapy (tamoxifen and aromatase inhibitors), (b) targeted therapy (herceptin), and (c) standard chemotherapy (doxorubicin and taxol). In preliminary experiments we will extend AT to all forms of breast cancer therapy. Intensive imaging and pre and post therapy molecular studies will allow assessment of tumor vascularity and cell function in response to evolutionary therapy. To improve AT we will examine mechanisms to exploit the cost of resistance to both detect resistant populations and suppress their proliferation. Preliminary data has shown that breast cancer cell lines expressing P-gp pumps have accelerated metabolism and are more sensitive to glucose deprivation than their parental cell line. These cells also show increased energy depletion in presence of P-gp substrates, such as Verapamil, Cyclosporin A, Quinidine and Clarithromycin at sub-toxic levels, making these candidates for fake drugs, to be used to increase the cost of resistance of P-gp cells between periods of chemotherapy. P-gp expressing cells also show increased glucose uptake in presence of fake drugs. Finally, we will also study how the addition of estrogen during therapy intervals can be used to increase the fitness of ER+ cells and thus maintain the number of hormonal therapy- sensitive cells.